From 815e83952eec7bec6f05e61095e2f5578e208519 Mon Sep 17 00:00:00 2001
From: 0xbyt4 <35742124+0xbyt4@users.noreply.github.com>
Date: Wed, 11 Mar 2026 21:43:10 +0300
Subject: [PATCH] fix: prevent double TTS on Web UI voice messages

When voice mode is enabled and user sends a voice message on Web UI,
both the base adapter auto-TTS (play_audio) and the gateway voice reply
(send_voice) would fire, causing duplicate audio playback. Skip the
gateway voice reply for Web platform voice input since base adapter
already handles it.
---
 gateway/run.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/gateway/run.py b/gateway/run.py
index 511ccf51..2558c2af 100644
--- a/gateway/run.py
+++ b/gateway/run.py
@@ -1635,8 +1635,12 @@ class GatewayRunner:
                     )
                     for msg in agent_messages
                 )
-                logger.info("Voice reply: has_agent_tts=%s, calling _send_voice_reply", has_agent_tts)
-                if not has_agent_tts:
+                # Skip if voice input on Web platform — base adapter auto-TTS
+                # already sent play_audio, so sending another would be double.
+                is_web = (source.platform == Platform.WEB)
+                skip_double = is_web and is_voice_input
+                logger.info("Voice reply: has_agent_tts=%s, skip_double=%s, calling _send_voice_reply", has_agent_tts, skip_double)
+                if not has_agent_tts and not skip_double:
                     await self._send_voice_reply(event, response)
 
             return response